In un-trusted environments such as where computing devices are connected to a network such as the Internet, some applications or services desire to identify particular computing devices, to verify communications as originating from the particular computing device, and to relate the particular computing device to a group of computing devices that share an account or are “tenants” of the service. Existing systems fail to disclose or suggest an automated mechanism for implementing such applications or services. For example, some existing systems use unsecured electronic mail as a means of authenticating data from a computing device. In such systems, the service has to simply trust that the electronic mail originated from the stated computing device and has not been corrupted or tampered with.